[Razor Blade]

Where did you hear that you had to reinstall Windows?! That's utter nonsense. All's you have to do is switch it off in youre bios. Actually today I had it on (heh, forgot I did!) And switched it off and Windows boots faster now. How odd

 

[WindHydra]

Hydra: Of course you didn't name what ATi card and what P4 he has. Not all P4 CPUs support HT, and not all ATi cards are capable of running D3 well at high settings.

It's a P4 3.2GHz and Radeon 9200.

 

[Kane]

And there lies the problem. The 9200 is a castrated 8500, the 8500 which was on a par with my geForce 3. The geForce 3 which ran Doom3 acceptably in mid detail at best.

 

[WindHydra]

I see... all these numberings are confusing. Why can't they use something different like TI and MX to differentiate the cards instead of just using FX and you have to tell by the number :???:
Ti and MX do stand for the better and the worse, right? oO

 

[industrian]

My MX card ran Doom 3 and Half Life 2. Albiet with low settings, but still better than most PS2 games.

I'm not sure about the Ti, but i believe that the MX440 is quite identical to a GeForce 2.

Correct me if I'm wrong.

 

[Kane]

It's ATi who really have the confusing numbering system:
9100 = 8500
9100SE = castrated 9100
9200 = castrated 9100
Then you have all the Pro/XT/SE variants of all the cards.
The GeForce 4 MX has a Dx7 feature set, and is faster than, while having the same feature set as, the geForce 2s.
The GeForce 4 Ti has a Dx 8.0 feature set
The 5xxx series is Dx 9.0 ranging from slow (5200) to fast (5950)
The 6xxx series adds the LE (Slow) GT (faster) and Ultra (fastest) monnikers to allow more products. All 6xxx cards are Dx 9.0c parts.
The ATi naming to my mind is confusing. An XT or a Pro both sound good, but IIRC one or other of them is the slower part.... Ultra sounds like the dogs bollox

 

[Sampah]

From what I've seen in benchmarks. if something doesn't support Hyper threading...the program will slow down...and AMD always wins in gaming, while intel has stuff like video and sounds edit software...I personally would get an AMD for many reasons (speed, cost, and how well heat is dealt with)....Where Intel wins in speed amd isn't far behind....

yup, I agree AMD always wins in gaming, I saw many benchmark using games and well of course AMD always rules. and the most important thing AMD processor is cheaper than intel.

 

[Badaro]

It's ATi who really have the confusing numbering system:
9100 = 8500
9100SE = castrated 9100
9200 = castrated 9100

AFAIK 9200 == 9100 w/AGP8x

the math program is a lab project so I can't even share the name of it.

Single or multi-threaded?
Float or integer math?
Which compiler?
Optimization settings for the compiler?

[]s Badaro

 

[ChrisRay]

AFAIK 9200 == 9100 w/AGP8x



Single or multi-threaded?
Float or integer math?
Which compiler?
Optimization settings for the compiler?

[]s Badaro

No. A 9100 is an 8500 with slower clock speeds, (4x2 card). A 9200 is a slower 9000 Pro. (4x1 Card)

 

[cooliscool]

Yep.. the 9100 has AGP8X.

 

[WindHydra]

What the... you mean the higher number is slower than the lower number??! oO That's so messed up.

 

[ChankastRules]

Single or multi-threaded?
Float or integer math?
Which compiler?
Optimization settings for the compiler?

Single-threaded
float math ( some integer as well but float would be the heavy part )
MS VC 2003 ( optimized for x86 , not intel or AMD specific )

 

[Badaro]

What the... you mean the higher number is slower than the lower number??! oO That's so messed up.

The number is probably how long it takes to run the benchmark.

No. A 9100 is an 8500 with slower clock speeds, (4x2 card). A 9200 is a slower 9000 Pro. (4x1 Card)

Ah, my mistake, I confused the 9100 with the 9000.
You're right, 8500 = 9100 > 9000 = 9200 (at the same clock).

Single-threaded
float math ( some integer as well but float would be the heavy part )
MS VC 2003 ( optimized for x86 , not intel or AMD specific )

That's probably why you're getting better results with the AMD.

[]s Badaro

 

[ChankastRules]

The number is probably how long it takes to run the benchmark.

Yes, ms is milleseconds :O, it's not a score... it really makes more sense to see how long it takes to run computations, since that's what is interesting.. atleast assuming both computer calculates everything correctly LOL!

That's probably why you're getting better results with the AMD.

Yep, as you can see the P3 is much better at running many things at the same time, while the AMD is better in the single-thread case for this scenario of computations, which is in the first place why I'm dissapointed I got an intel , and is why I started this thread

 

[Badaro]

Yep, as you can see the P3 is much better at running many things at the same time, while the AMD is better in the single-thread case for this scenario of computations, which is in the first place why I'm dissapointed I got an intel , and is why I started this thread

I'm no expert on the topic, but I don't think so. HT allows for better processor usage when running multiple threads, but shouldn't cause any problems with single-threading applications.

My guess is that the large differences are caused by you using floats and non-optimized builds. Try running a P4/SSE2 optmized build versus an AMD optimized build.

[]s Badaro

 

[D.D.]

it wasnt AMD optimized. it was optimized for no one.

 

[WindHydra]

But for single thread programs does HT really only use half of the CPU power?

 

[Badaro]

it wasnt AMD optimized. it was optimized for no one.

I didn't say that. I said that a float-based math program, in a non-optimized build will probably be faster in an AMD, because AMDs have a better FPU.

In other words, the compiler isn't favoring a specific processor, but the program itself is. However, a good compiler could reduce those differences by using some advanced processor features, such as SSE/SSE2.

But for single thread programs does HT really only use half of the CPU power?

No, that's a display issue with the Windows Task Manager.

[]s Badaro

 

[D.D.]

I didn't say that. I said that a float-based math program, in a non-optimized build will probably be faster in an AMD, because AMDs have a better FPU.

oh i c wut ur talking aobut now XD. by "Try running a P4/SSE2 optmized build versus an AMD optimized build." i thot u mean "instead of using an AMD optimized build, use a P4 optimized build" .

 

[ChankastRules]

No, that's a display issue with the Windows Task Manager.

Well, if you run two threads ( the programs at the same time ) you can see that the increase it takes to calculate the result is not much longer, compared to running them separated, that's why I say it doesn't use much more than 50% of it's potential computational power on process such as the benchmark I sent.